It is currently believed that a high percentage of human cancer results from exposure to environmental chemicals. Little is known about the abilities of the environmental chemicals which are carcinogens in experimental animals to produce tumors in humans. Part of this uncertainty results from the lack of suitable data on which to base extrapolations from the high doses used in animal studies to the low levels found in the environment. Differential response among species is a compounding problem. As most carcinogens are believed to initiate carcinogenesis via a mutation-producing interactions with DNA, knowledge of the types of DNA adducts which lead to mutations and the mechanisms by which the adducts are formed or repaired could be important in understanding factors associated with risk. In this proposal several methods of identifying premutagenic-DNA adducts produced by N-nitroso compounds and simple alkylating agents in Salmonella typhimurium will be employed. These involve the correlations of particular adducts with mutations assayed both in forward and reversion assays and are: 1. identification of adducts formed with the same dose response as mutants, 2. identification of adducts formed at similar levels by equimutagenic doses of different mutagens which generate the same adducts. 3. investigation of the effects of different bacterial DNA repair systems on mutations and adduct levels and 4. treatment of bacteria with mutagens, incubation or outgrowth in liquid suspension, and determination of mutants and DNA adducts before and after the incubation. Also mutagens will be reacted with naked DNA and adducts compared with those obtained above, to determine the effects of permeation or rapid repair on adduct levels. Results of all of the above experiments will be used to establish a potency index of premutagenic adducts and to dissect mutagenesis into its component processes. In addition, mutagenesis and data on in vitro metabolism of a large group of nitrosamines will be determined and expressed in such a manner that correlations between mutagenic and carcinogenic potencies (if they exist) may become evident. Additional studies will attempt to characterize some of the bacterial DNA repair systems with respect ot substrates, cofactors and specificities.